388 Transact ions . — Chemistrij. 



(8.) That clay is similarly affected by hot paraffin. 



(9.) That clay can readily be charged with some of the constituents of 

 petroleum, to such an extent as to have almost the consistence as well 

 as the appearance of torbanite. 



These results have, I believe, an important bearing upon our present 

 theories as to the formation and constitution of the valuable mineral torbanite. 

 As to the formation of this mineral, it plainly appears from them that clay 

 strata can abstract the colouring matter of petroleum. If this process is 

 carried on to a small extent we have only a feebly bituminous clay, but if 

 carried on till the clay is saturated, or nearly so, we have a mineral which I 

 believe has exactly the constitution of torbanite. During the formation of 

 this mineral the petroleum passing through it would be purified to a greater 

 or lesser extent. 



From what has been already stated, I feel sure the absorption is not of a 

 mechanical but of a chemical nature, and this brings me to the next point, 

 that is, the nature of the mineral in question, torbanite or bog-head coal. 



As to its constitution, this mineral is associated with the amber group in 

 our best mineralogical works, and the earthy matter is thrown out of the 

 formula. Now, this earthy matter is within small limits uniform in amount 

 in the case of all the samples of this mineral yet analysed, being from nineteen 

 to twenty-six per cent., and it is essentially silicate of alumina, that is 

 anhydrous clay. 



I Consider, therefore, the ash of this mineral is not an accidental element 

 as now stated, but is an essential part of it, — that, in fact, torbanite is a 

 combination of a bituminous kind of substance with clay, the water of the clay 

 being either substituted by it or a bitumino-silicate of alumina formed, which 

 substance may have no affinity or but a very sliglit one for water. 



The colouring matters for petroleum and kerosene are in general 

 terms described as of a bituminous nature — bnt whether bitumen itself is 

 actually or universally present in them has not yet been demonstrated. 

 However, these colouring matters are certainly oxidized hydrocarbons, and so 

 class with bitumen and the combustible part of torbanite. 



Being thus oxidized hydrocarbons they can hardly fail to be of an acidic 

 nature, and so the statement as to their capability to chemically combine with 

 clays as shown, is one which a consideration of the basic nature of clay will, I 

 think, greatly predispose us to admit as a correct one. 



In conclusion, these results tend to indicate — 



(1.) A cheap and expeditious method for purifying our coloured kerosenes, 



one in which there need hardly be any waste. 

 (2.) That by using pure clay useful pigments may perhaps be obtained in 

 this manner. 



